Original Article Diagnostic value of bone-specific alkaline phosphatase metastases: a meta-analysis
Int J Clin Exp Med 2015;8(10):17271-17280
www.ijcem.com /ISSN:1940-5901/IJCEM0012164
Original Article
Diagnostic value of bone-specic alkaline phosphatase
in lung carcinoma patients with bone
metastases: a meta-analysis
Qing-Tao Zhao1, Zhao-Xu Yang2*, Lei Yang3*, Dong Xing2, Jing-Chao Wei2, Wen-Yi Li2
Departments of 1Thoracic Surgery, 2Orthopedics, Hebei General Hospital, Shijiazhuang 050051, Hebei, P. R.
China; 3Department of Pediatrics, Bethune International Peace Hospital of Chinese PLA, Shijiazhuang 050082,
Hebei, P. R. China. *Co-rst authors.
Received June 30, 2015; Accepted October 9, 2015; Epub October 15, 2015; Published October 30, 2015
Abstract: Aim and Backgrounds: The accurate diagnosis of lung carcinoma patients with bone metastases is crucial
for therapy and the prevention of complications. We performed a systematic review and meta-analysis to evaluate
the diagnostic value of serum bone-specic alkaline phosphatase (BALP) in lung carcinoma patients with bone me-
tastases. Methods: Such databases as PubMed, Embase, Cochrane Library, Web of Science, Ovid, BioMed Central,
Biosis previews and four Chinese databases (Chinese Biomedical Literature Database-disc (CBM), Chinese National
Knowledge Infrastructure (CNKI), Technology of Chongqing (VIP) and Wan Fang DATA) were retrieved on computer,
and the relevant journals were also manually searched to collect the trials on BALP in diagnosis of lung carcinoma
patients with bone metastases. The meta-analysis was conducted by using Meta-Disc 1.4 software. Results: A total
of 8 studies were included, and there were 848 lung carcinoma patients diagnosed by gold standard, patients were
divided into two groups: 419 cases with bone metastases and 429 cases without bone metastases. The meta-
analysis showed that, the pooled sensitivity (SEN), specicity (SPE), positive likelihood ratio (PLR), negative likeli-
hood ratio (NLR) and diagnostic odds ratio (DOR) was 0.48 [95% CI (0.43 to 0.53)], 0.86 [95% CI (0.82 to 0.89)],
3.14 [95% CI (2.47 to 3.99)], 0.62 [95% CI (0.56 to 0.68)], 6.66 [95% CI (4.62 to 9.60)] respectively. And the AUC
of SROC was 0.78, (Q*=0.72). Conclusion: BALP has greater diagnostic value in detecting lung carcinoma patients
with bone metastases. However, further large scale studies are required to conrm the predictive value.
Keywords: Lung carcinoma, bone metastases, BALP, meta-analysis
Introduction
As the second leading cancer type for the esti-
mated new cancer cases, lung carcinoma rep-
resents the major cause of cancer death in
both females and males [1]. Bone metastasis
can be found frequently in lung carcinoma [1,
2]. It is reported as 24-40% in clinical studies
and 36-40% in autopsy series [3, 4]. Lung car-
cinoma frequently develops bone metastases
in advanced stages of disease [5]. The main
symptoms of bone metastasis include severe
pain, pathological fractures, spinal cord com-
pression, hypercalcemia, anemia and so on [6,
7]. But up to 20-25% of patients are asymptom-
atic [7]. These skeletal-related events have
been associated with reduced quality of life
and reduced overall and median survival, so
the early diagnose of bone metastasis and
effective therapy could be initiated timely and
improvement of life quality and treatment to the
patients may be achieved [8, 9].
Diagnosis of bone metastasis is usually per-
formed initially with plain radiography or com-
puted tomography (CT) or magnetic resonan-
ce imaging (MRI) or bone scintigraphy screen-
ing and conrmed by whole body bone scan by
single-photon emission computed tomography
(SPECT) [10-12]. However, they have very low
sensitivity in detecting bone micro metastasis
[12]. Bone scan is excellent for whole-body
screening and can detect micro metastasis of
bone metastasis [13, 14]. However, it can give
false-negative results in lytic bone lesions and
the risk of radioisotope exposure. Due to SPECT
have high price and radioactivity, it is not a nec-
essary recommendation for newly diagnosed
patients [14].
Diagnostic value of BALP in LC patients with BM: a meta-analysis
17272 Int J Clin Exp Med 2015;8(10):17271-17280
In contrast, the detection of serum bone meta-
bolic markers is cheap and easy to perform,
and may assist in the early diagnosis and as-
sessment of therapeutic results in bone metas-
tasis [15-17]. BALP is the bone-specic isoform
of alkaline phosphatase, which originates from
many tissues, but primarily the liver and bone
[18, 19]. BALP is a tetrameric glycoprotein
found on the cell surfaces of osteoblasts [18,
19]. The exact function of BALP remains unk-
nown. However, it has been suggested that it
might play a role in mineralization of newly
formed bone [15].
There have been studies reporting the use of
serum BALP as a serum marker for bone metas-
tases in patients with lung carcinoma, but the
results are heterogeneous and even conicting
[20-22]. The practical value of these markers
has yet to be fully evaluated. The objective of
the present review was to synthesize and ana-
lyze the results from systematic selection of
research papers that evaluated the diagnostic
accuracy of serum BALP by directly diagnosis of
bone metastasis in patients with lung carci-
noma.
Methods
Search strategy
A comprehensive systematic literature review
of original researches studying the diagnostic
accuracy test accuracy of BALP in lung carcino-
ma patients with bone metastases was per-
formed searching the following electronic data-
bases through February 15th 2015: PubMed,
Embase, Cochrane library, Web of science, Ov-
id, BioMed central, Biosis previews and four
Chinese databases (CBM, CNKI, VIP and Wan
fang DATA). In addition we conducted supple-
mentary searches in the references of the
retrieved articles. Titles and abstracts were
reviewed for relevance. Relevant prospective or
retrospective cohort or case-control studies
were included in the meta-analysis. Subject
headings and keywords used in the search pro-
cess included the following: “bone-specic al-
Figure1. The study selection and inclusion process.
Diagnostic value of BALP in LC patients with BM: a meta-analysis
17273 Int J Clin Exp Med 2015;8(10):17271-17280
kaline phosphatase”, “BAP”, “BALP”, “BSAP”,
“sBAP”, “lung cancer”, “lung carcinoma” and
“lung neoplasms”. The controlled vocabulary
search terms for different databases are not
identical. Therefore, search strategies need to
be customized for each database.
ents, and reference test for the analysis of SEN
and SPE (the number of true positive (TP), false
negative (FN), true negative (TN) and false posi-
tive (FP) results) for comparison of lung carci-
noma patients diagnosed with bone metasta-
ses vs. control. Any disagreements were re-
Table 1. Summary of the diagnostic results of the included studies
Study Year Country Assay
method NO. TP FP FN TN SEN SPE
Aruga A 1997 Japan EIA 91 28 3 35 41 44.4% 93.2%
Alatas F 2002 Turkey EIA 52 24 14 311 89% 44%
Ebert W 2004 Germany CLIA 138 11 0 38 89 22% 100%
Kong QQ 2007 China ECLI 96 22 4 39 31 36.1% 88.6%
Lumachi F 2011 Italy ELISA 35 6 3 10 16 37.5% 84.2%
Bayrak SB 2012 Turkey ELISA 65 7 4 16 38 30.34% 90.48%
Tang C 2013 China ELISA 265 82 31 48 104 63.1% 77%
Xin Y 2010 China ECLI 90 20 2 30 38 40% 95%
EIA: Enzyme immunoassay; CLIA: Chemiluminescence immunoassay; ECLI: Electro-
chemiluminescence immunoassay; ELISA: Enzyme-linked immunosorbent assay; TP:
True positive; FP: False positive; TN: True negative; FN: False negative; SEN: Sensitiv-
ity; SPE: Specicity.
Inclusion criteria
Studies were considered eli-
gible for inclusion if they met
the following criteria: I) Study
design. Observational studies
(cohort or case-control stud-
ies). II) Population. Lung carci-
noma patients with bone me-
tastases, or without bone me-
tastases. III) Diagnostic test.
Serum BALP in lung carcino-
ma patients. IV) Reference te-
st. The following reference tes-
ts were considered eligible:
radiologic examination (X-ray,
CT, MRI), histological exami-
nation, etc.
Exclusion criteria
Studies were excluded from
the meta-analysis for the fol-
lowing reasons: I) Duplicate
publication; II) No human stu-
dies; III) Necessary data could
not be obtained.
Study selection
All the studies were review-
ed by two reviewers indepen-
dently based on titles and
abstracts, and then the full te-
xts of potentially eligible stud-
ies were retrieved for furth-
er assessment. We resolved
disagreements by reaching a
consensus through discussi-
on.
Data abstraction
The following data was ex-
tracted from the included st-
udies by two reviewers inde-
pendently: authors, year of pu-
blication, journal, study desi-
gn, number of eligible pati-
Figure 2. Presentation of QUADAS-2 results.
Diagnostic value of BALP in LC patients with BM: a meta-analysis
17 274 Int J Clin Exp Med 2015;8(10):17271-17280
solved through consultation with the third re-
viewer.
Assessment of methodological quality
The methodological quality of the included
studies was independently assessed by two
authors, using the Quality Assessment of Dia-
gnostic Accuracy Studies 2 (QUADAS-2) tool
[23], which consists of four domains: patient
selection, index test, reference standard, and
ow and timing. QUADAS-2 is an updated ver-
sion of this evidence-based quality tool. All
domains are assessed for risk of bias and the
rst three domains are assessed for applicabil-
ity by indicating a “low”, “unclear”, or “high” rat-
ing. This tool helps to evaluate the principal
methodological risk of bias in systematic revi-
ews of diagnostic test accuracy [24]. Specic
coding instructions adapted for this review will
be included for the reviewers. In case of doubt,
a third and fourth reviewer were consulted.
Data analysis
Standard methods recommended for meta-
analysis of diagnostic accuracy were used. The
number of TP, TN, FP and FN were retrieved
from each article by two investigators indepen-
dently and entered into an excel datasheet [24,
25]. Discordant ndings were assessed in a
joint approach and authors asked for verica-
study weight [24, 27]. The area under the AUC
represents an analytical summary of the test
performance and illustrates the trade-off be-
tween SEN and SPE [24, 27]. The chi-square-
based Q test and the inconsistency index I2
were used to detect statistically signicant het-
erogeneity across studies. When a signicant Q
test (P<0.05 or I2>50%) indicated heterogene-
ity among studies, the random-effect model
(DerSimonian-Laird method) was conducted for
the meta-analysis to calculate the pooled SEN,
SPE, and other related indexes of the studies;
Otherwise, the xed-effect model (Mantel-Ha-
enszel method) was chosen. Chi-square test
was used to detect statistically signicant het-
erogeneity across studies. Additionally, we also
calculated the Spearman correlation coef-
cients. A strongly positive rank-correlation coef-
cient and a value of, 0.05 are indicative of a
signicant threshold effect.
Results
Search results
A total of 278 titles and abstracts were prelimi-
narily reviewed, of which 8 studies were avail-
able for the meta-analysis, including 848 lung
carcinoma patients who received serum BALP
tests [21, 22, 28-33]. Figure 1 shows a ow dia-
gram of the selection process. The characteris-
tics of each study are shown in Table 1.
Figure 3. Summary receiver operating characteristic (SROC) curve for BALP
in the diagnosis of Lung carcinoma patients with bone metastases in the 8
included studies.
tion when in doubt. The Meta-
DiSc 1.4 (XI Cochrane Collo-
quium, Barcelona, Spain) was
used to perform all data anal-
ysis. The following indexes of
test accuracy were comput-
ed for each study: SEN, SPE,
PLR, NLR, DOR and genera-
te the bivariate SROC curve
[25]. The DOR value ranges
from 0 to innity, with higher
values indicating higher accu-
racy levels [26]. Data were
presented as forest plots and
receiver operating character-
istic curves. Forest plots dis-
play the SEN and SPE of indi-
vidual studies with the cor-
responding 95% condence
intervals. The receiver operat-
ing characteristic curves sh-
ow individual study data po-
ints with size proportional to
Diagnostic value of BALP in LC patients with BM: a meta-analysis
17275 Int J Clin Exp Med 2015;8(10):17271-17280
Assessment of methodological quality
When using the QUADAS-2 tool to review the
eight included articles, it was determined that
three studies [22, 29, 32] had low risk of bias
and low concern regarding applicability. Three
studies [21, 28, 33] were found to be at risk for
bias, but had low concerns regarding applicabil-
ity. The nal two studies [30, 31] were judged to
be at risk of bias and as having concerns
regarding applicability (Figure 2).
The SROC
The corresponding SROC (Figure 3) shows an
AUC of 0.78 with standard error=0.02, and the
pooled diagnostic accuracy (Q*) was 0.72 with
standard error=0.02, indicating high overall
accuracy of BALP for the diagnosis of lung car-
cinoma patients with bone metastases.
The Spearman rank correlation coefcient was
0.64 (P=0.09), conrming that the variability
across these studies could not be explained by
differences in the diagnostic threshold.
The pooled DOR
Signicant heterogeneity among the studies
was not detected (Cochran Q statistic=5.38;
P=0.61). A Forest plot for the DOR of BALP for
the diagnosis of lung carcinoma patients with
bone metastases was 6.66 with a correspond-
ing 95% CI of 4.62-9.60, as shown in Figure 4.
Figure 4. Forest plot for the diagnostic odds ratio (DOR) of BALP to diagnose Lung carcinoma patients with bone
metastases. DOR (diagnostic odds ratio)=6.66 (95% CI, 4.62-9.60).
Figure 5. Forest plot for the sensitivity of BALP to diagnose Lung carcinoma patients with bone metastases. Sensitiv-
ity=0.48; (95% CI, 0.43-0.53).
6
7
8
9
10
1
/
10
100%
